Better and worse ways to spot a liar
bbc.com
bbc.com
It strikes me to the degree that the folklore is mixed with the (so called dubious) science. I was always interested in the topic, and had indeed read a book when I was teenager, with ideas such as the iconic example of Clinton touching his nose.
About 5 years ago I went to the topic again; learned that one of the biggest authority in the topic was Paul Ekman and read a couple of his books
Surprise surprise, the main takeaways were ideas such as:
(...) not to jump to early conclusions: just because someone looks nervous, or struggles to remember a crucial detail, does not mean they are guilty. Instead, you should be looking for more general inconsistencies.
Or
There is no fool-proof form of lie detection, but using a little tact, intelligence, and persuasion, you can hope that eventually, the truth will out.
Ekman repeated all over the place that there is no body language for lies, only for emotions, and that the emotions can have a variety of causes. And that was already clear the last century! If some security entity has bought something that promised to spot lies, it was probably folklore-based and no science-based.
It is so sad that nowdays it is not seen as absurdal that some kind of policeman is asking passangers about their travel plans. The journalist get excited that new methods of catching "cheating passangers" are beind developed.
Apparently in the brave new World we have created this is considered normal.
"[B]efore 1915 His Majesty's Government did not require a passport for departure, nor did any European state require one for admittance except the two notoriously backward and neurotic countries of Russia and the Ottoman Empire."
And it wasn't until 1945 that Americans were generally required to hold a passport to travel abroad during peacetime. [1]
An american in 1940 may not have needed a passport to leave, but would they have needed one to get in somewhere else? Surely the nations of europe at that time would have been suspicious of basically everyone?
People used to laugh at the Iron Curtain countries for asking such questions...
In the 60s you could board a plane without even passing any metal detector...
If the incident leading to the condition happened recently, then that would be an excuse for not knowing sign language. Yes, they may still need to write their answers, but it is universally accepted that most people don't like to write so they could keep their answers brief without raising suspicion. By writing slowly, they may even drive the interviewer to the point they cut the questioning short. Very serious bad guys could even get a botox injection beforehand, paralyzing their lips or get their mouth wired shut.
My point is that at the end of the day, the goal here is to catch very serious bad guys. Any truly dangerous person with the intent and actual capability to do something very bad should have tools of deceit on par with their capabilities for harm up to and including getting one's mouth wired shut. Note that for this discussion I'm discounting "shoe bomber" types as not really having a capability for harm. I suppose that at the end of the day, that's who techniques like these are aimed at in the first place. Alas, there is no such thing as perfect security.
I'm not sure if the goal is solely to stop "truly bad guys" who want to bring down a plane from getting on it, but rather make a note of anyone who might be traveling under false pretenses for any number of reasons.
This is what I'm afraid of. Is travel on a plane now squarely reserved for the honest or those with nothing to hide? There are plenty of very legitimate reasons to travel under false pretenses, including it would seem, hiding from one's own government because of some irrational administrator who decides to use air travel restrictions to carry out a grudge. The restriction on travel, especially within the US as applied to US citizens is very disturbing as it represents a fundamental erosion of our freedoms. Take a look at this case as just one example:
https://www.eff.org/deeplinks/2015/04/united-airlines-stops-...
Not at all. They'll still let you on the plane after grilling you. They just want to know that you're hiding something. For future reference.
Considering several horror stories and uneeded questioning (of students etc) entering the country, not that "good" really.
...which doesn't imply that a "just question 100% of the people" strategy would work—they need to question few enough people that the planes still leave on time and the airlines stay in business. But within that limit, it's in their best interest to be as trigger-happy with the detainments as possible.
How do you know how good they are? That seems like the point of the article.
After almost a year of arguing with him, he offered to transfer the domain before the end of the "contract". But for "technical reasons" it had to be done after the hosting company had shut down the web site and e-mail, which caused some downtime for us. In retrospect it would have been much easier if I hadn't questioned him in the first time.
If you want to get someone to do something then yes you have to use different techniques (allow them to save face etc.).
Border agents have the law, manpower, firepower, and territory on their side and are ready to use it at a moment's notice.
Edit: Don't get me wrong, he was a pathological liar as well ;)
Perhaps questioning him too and him feeling he was being pushed too far broke his implicit sense of trust, feeling loss of favourability.
It may have been a verbal agreement. It may have been his interpretation / perception of a verbal agreement (warped by assumptions, or by time), or it may have been a lie. Hard to tell.
A nice simple example that I'll remember, thanks.
Only then do you end the contract. (Kind of scary that the org didn't have a copy of his contract on file, though. Sounds disorganized.)
Agree almost never the right thing to do. Trapping people makes them dig in their heels typically as opposed to giving them a graceful way to give you what you want. All of this is situation specific of course but as a rule I never call people out on their lies in negotiation.
"Once I had to work with a freelance web master and we were trying to cancel his contract and make him transfer the domain name to us (an association I was working for)."
Honestly, best way to handle this type of things is to go the "scary lawyer letter" route. Try nicely (without threatening) and if that fails get an attorney to write the "scary lawyer letter". At that point the idea of actually getting some money (rather than having to hire an attorney to defend a verbal contract) would seem very attractive. There are other methods that can be employed (UDRP, trademark law) however they are more expensive and depend on the specifics.
Big difference.
That again is a different problem: one of communicating your findings to others.
Oh the irony surrounding statistics and how often it is used to lie - case in point: the mode is completely absent and some useless factoid is presented instead.
That's not to say that argument is false ("you can't use body language"); merely that because of the useless information we don't know if it's true either.
My first thought is that with 20,000 participants, sheer chance will give you a handful of outliers, and 50 at 80% accuracy doesn't sound very high unless there were a very large number of trials per participant.
https://en.wikipedia.org/wiki/Wizards_Project#Synopsis
Granhag, Pär; Strömwall, Leid (2004). The Detection of Deception in Forensic Contexts.
The actual paper was linked lower down: https://news.ycombinator.com/item?id=10180932
Friday night: go out with quiet friends. Saturday AM questioning: "How was your evening?" "Fine". "OK".
Saturday night, go out with less quiet friends. :) The Sunday AM questioning was rather more rigorous.
So I described what I did Friday. There were telling hesitations, of course. But 90% of the questions had immediate and honest answers. Just with the day changed.
The first one is to offer to whoever is questioning you a "local maximum" victory, i.e. to pretend to hide a smaller lie compared to the bigger one you're supposed to be hiding. In the case you mentioned, you could pretend to be "covering up" the fact that you've been cheating on your SO, hence all the non-concordances related to your past locations. When the guys questioning you finally realize that you're lying on something, they'd think you had been doing so out of fear of not being seen as an unfaithful spouse, when in reality you want to blow things up (just to give an example). The downside to this strategy is that this fails when the "admission criteria" is 0 and 1, you're either a liar, which means you're out, or not, which means you're in. The one mitigating factor is that almost everybody lies all the time.
The second strategy is to be so good at this as to believe in your lies/stories you're telling. This is a little sociopath bordering on schizophrenic, but I've seen some people doing it almost to perfection (granted, the stakes were not very high). The one weak point of this is that you have to know exactly when and how to push the on/off switch.
Now that I think of it, I see that happen quite a bit at the poker table too. You can draw chips out of other players by making them think that you're just trying to bluff, but you really have a good hand.
One of my favorite quips from the internet is:
All really good salesmen temporarily believe
whatever bullshit they are selling at the time.
It's kind of like method acting.
That is harder than it sounds. Even sociopaths have a hard time outright lying. Or keeping lies straight.
The technique being used here is one of consistency. Is your story consistent? Does it have all of the details one expects from a normal story? Are there any hesitations when recalling the story?
It is very difficult to come up with consistent and detailed lies on a moments notice. The better approach (used by lying CEOs on news shows) is to have a different story that you tell in detail.
Do all your imagining ahead of time. Walk through and construct the lie. Fill in the details. You're not just imagining this one thing, you're imagining an entire alternate reality where this thing is true and all the implications that follow from that.
When it comes time to answer questions the person is asking about the alternate reality you've imagined. No need to make things up - just recall back.
I guess I'm saying the same thing as you, just in a different way - don't make something up on a moments notice, just have a different story that you tell.
- Why were you there? - Vacation. - How long were you there? - 2 weeks. - Who did you go with? - Him (points to brother)
Treating people like idiots, and while asking dumb questions, doesn't seem to be the best way to spot a liar. I'm glad the science of it is improving.
They're still interesting to look out for, though, as they're helpful hints to let you direct your conversation to probe at areas that makes someone nervous and/or to figure out what someones different tells means.
That's fine if you believe that, but it is possible to be a winning live player and not use tells. Tells being used are more anecdotes than typical situations that come up.
At live poker, the number of hands ranges from 15 to 30 per hour. Most hands don't go to showdown, so you simply never get the information you need. Unless you are playing higher stakes with a smaller player pool, or a small home game, you may not have the same opponent at your table for a long time even if they're a regular.
Giving off tells assumes that opponents are perceptive enough to notice, or naive enough to fall for overly obviously tells. This is asking too much for a lot of weaker opponents. And most people won't be involved in the hand and paying attention.
The most useful "tells" may not even qualify as such. They're usually weak or bad players trying their hand at bad acting in a way that tends to be obvious after you've played enough.
You also run into the problem that even amateurs can familiarize themselves with tells and try giving off "reverse tells" which reduces it to a game of "leveling".
Most of the time the best decision is irrelevant of whatever behavior an opponent is perceived to be exhibiting. If I'm following through on a bluff against an opponent on a flush draw (comes 35% of the time if they see the turn and river), and their draw comes in and they sigh "guess I'm beat" and shove into me on the river, I can't call with ten-high.
Still, the 'big lies' are just getting stronger and these techniques are useless on them. How many political identities can be dismantled by 'surprise'?
Maybe it is just me that is sweatier and wilier when stressed.
TL;DR
Use the deviation from the base-line to work out if someone is lying.
So my reading is 2/3 mock passengers identified as true positives, 1/1000 real passengers identified as false positives.
(With some provisos as noted in the paper)
The 'active' method in the article is useful, but has a limitation: you need to be able to ask questions in real-time.
Collating the answers and automating truth evaluation would be a pretty interesting AI problem. It should also be possible to have an AI formulate the statistically optimum questions to ask, Akinator style.
P.s. That would be the killer app for Google Glass, right there. Would also increase chance of being thrown out of pubs by 1000%, but hey.
Which is to say this probably catches confused people and people hiding harmless but embarrassing facts but probably isn't useful against "determined evil doers".
I also wondered why where I was coming from had any relevance to the speed I was currently driving but always sort of figured it was some kind of fishing technique.
Deleted comment
Read to me less like academic research and more like a before/after show by the training consultant. But I only read if very fast.
http://www.huffingtonpost.com/alan-carniol/inside-the-star-i...
> Even if we think we have a poker face, we might still give away tiny flickers of movement known as “micro-expressions” that might give the game away.
but then goes on to say: > ... this too seems to have been disproved.
You can spot liars based on body language, but then you first need to know how that specific person reacts in normal situations vs. when they're lying.
Yet juries in rape trials supposedly do it all the time...or don't they?
It is also possible that both victim and rapist are telling the truth, but it still is an open question if it's a rape. Example: Julian Assange case with the broken condom.
The general technique is probably effective though.
I am really uncomfortable in front of people, and no matter how honest I am, I am show my discomfort in interacting with people quite a lot. I look around the room while talking to people, I shuffle around, sometimes I sweat just talking about the weather... I have social anxiety obviously, but it's awkward to start a conversation with anyone by saying, "Oh hey, this interaction is going to be really awkward because I have social anxiety." I tried it for a while and most people seemed to just blow it off.
It's caused me quite a few issues with friends who frequently think I'm lying when asked 'truth-seeking questions. More importantly authority figures tend to misinterpret this as me being deceptive or uncooperative.
---
The most recent example was a police officer that came to my house to see if I had seen my neighbor's car recently. The car had been stolen and they were trying to determine the last known time it was present. As usual, during the rather normal questions I was rocking back and forth, chewing on my nails and shuffling my feet. I'm painfully aware of these things and consciously try to stop each little tick, one by one.
It didn't take long for the officer to ask why I was so nervous, and the he promptly switched the subject and asked if I had any hobbies. I'm fairly obsessive about my hobbies, and I pretty much immediately started rambling about what I was doing. I suspect I stopped most of my 'nervous ticks', because he interrupted my ramble to ask why I was lying about not having seen the car lately.
It was extremely offputting, since I wasn't lying. I got really nervous again, and started thinking about how I 'screwed up' the interaction. Instead of responding to his question I simply told him that I had really bad social anxiety and this questioning was really difficult for me. That didn't help at all.
I ended up with another officer at my door, with more questions that I had no answers to, and I became more and more nervous. Needless to say this went on for ~30 minutes just standing at my front door until I suppose they realized that I was either telling the truth, or was a completely unreliable witness (I was!).
---
Things like that aren't rare for me. It sucks, and early in my life it caused me to simply lie a lot. People almost always thought I was 'up to something' or 'not telling the true', so I would just go with it. I eventually learned the value of consistent honesty, but I am treated the same regardless.
It goes without saying that this being in my head during every single personal encounter causes me even more anxiety and unsuredness about my responses to someone.
edit: I noticed I actually started rocking back and forth and itching my head randomly while writing this post... bleh.
I'm sorry to hear how much trouble it causes you. It's a miserable experience.
I think that the irony is that if I wanted to lie, I could probably get away with it since I give off such mixed signals when saying _anything_. True or false, I just fall apart.
If you want to correspond further about this, feel free to drop me an email.
I guess they can tell the difference between a socially-anxious perfectionist stressing about giving correct answers and a liar stressing about giving false answers... or the airline is wasting their money by giving this "interview" but ignoring really sketchy answers. Or maybe I did not appear as suspicious as I thought.
Another time I casually broke eye contact with a bus security officer while he was rattling off prohibited items and searching my bag. I've noticed sometimes I subconsciously prefer to focus by looking down and maybe turning an ear toward someone talking about something detailed like this, but this guy seemed to interpret it as a sign that he'd just said something that was in my bag and shouldn't be. He repeated the question once and maybe twice, so this time I stared straight into his eyes as I answered. Maybe it looked like a liar over-compensating, but whatever, he was tossing my bag anyway.
I was deaf since birth. So at a very early age I picked up body-language, micro-expressions and of course lip-reading which were rather an integral way for me to communicate!
> "The problem is the huge variety of human behaviour – there is no universal dictionary of body language"
Um. Yes there is. Everyone uses body-language. Everyone uses their mouth, their eyes and their hands. Take in sign language. While it's not the same in every country, someone fluent in any sign language can understand BSL (Brazil Sign Language) or NZSL (New Zealand...), ASL etc. Why? Because sign language is the most literal thing you can think of. If I look at someone and point to them, then point to someone else, what do you suppose that means? The only issue is local dialect/slang which is easy enough to figure out.
I'd like this BBC article to try this on deaf people and see what the results would be. It would be extremely different. Even for people who just wear hearing aids: a frown, does not mean anger... it means they're trying to understand you. If that person misheard a previous question, but then didn't mishear it the second time then... are they lying? No.
For myself in particular, when I was trying to have conversation with people I had a few difficulties. For this, think of dyslexia, say the brains language processor. For some people with dyslexia an example sentence could look like: "I ___ to ___ shop ___ the ______ ____ ___ car". It's exactly the same for someone who is deaf. However, they need to be working that language processor in their head 300% capacity. Not only are you lip-reading, using sound from your hearing aids, you're factoring in context, location, the person talking to you, body-language and so on. So a deaf person, will then fill in those blanks in my above example and hope they got it right. Except, by that stage, more has been said and you're now trying to remember what was said just a few minutes ago. Then you're defeated.
However, if you watched my body language in ann airport you'd probably shoot me or whatever customs does. I'd be the ideal 'liar' that this BBC article refers to.
Since I got my cochlear implant, (I jumped from 4% hearing to something like 80% upwards) my world has grown incredibly. Not only do I have my previous skills, but I can now add verbal input into my once stressed language processor. It's incredible what I can pick up on. Now that I have that extra sense/input, I find that I can tell whether someone is not being truthful or honest. Another poster here said that "give them enough rope to hang themselves with" and that's very true. Someone rambling? Watch their hands. Someone straight to the point, confident, and uses no body language -- very confident of themselves. So simply throw them off. Does their attitude change? If it does, what does that mean? Context comes into play here, and customs simply don't have the time. Nor will Police.
Someone trying to explain the minute details of their drive to work, watch their eyes and see where they go (looks you in the eye, wall, phone?). Then stop. Who exactly remembers details that they've got no reason to remember? So they'll tell the short version, 'cos they have done it 1,000 times. Then if you're probed such as this article says... you'll end up getting anxious, and contradict yourself. "oh, maybe I did take Stuart Street...".
I am a firm believer that body language is really a good way to determine language nuances, even in different languages. It works. I've been friends with people who didn't know English, but I could communicate with them effectively enough. Giving someone a few weeks of body language training, is going to do squat. Getting experts, again I'm not sure -- have they ever had to rely on it? Perhaps they should wear headphones with whitenoise and interrogate people, with someone who is listening -- and compare notes.
I kind of feel like writing a blog post to refute this article, with proper examples etc. Would anyone be interested?
I apologise if I sound disjointed it's 3am in NZ right now, and I just had the need to go "no this is not quite right".
P.S. When I went to Singapore, a customs person glared at me and nodded to the guy with the gun and so I smiled and I said, "Hi! I hope you haven't had a horrible night so far -- hopefully my documentation is in order and you'll not have to deal with boring stuff!" and she went from >:{ to :-) and nodded to the gun guy walking behind me, who turned around back to his spot. I got all that from a split second glance. It's actually even easier for me now with my implant to do this sort of thing in case actual spoken communication is required.
EDIT: As per article, it is common sense -- but you need to know someone well enough to take judgement, which these guys have no time for. Speaking for myself, I learned over a long period of time to do that as quick as possible. Otherwise, I'd have been left to fail.
I asked someone who was studying ASL about the differences between sign languages. She pointed out that they do commonly have different signs which might be very abstract and non-literal, often because they arose in different schools for deaf people (or different regions with a very high incidence of deafness) in earlier eras.
However, some of the sign languages have some etymological relationship with one another, especially French Sign Language (LSF)
https://en.wikipedia.org/wiki/French_Sign_Language
https://en.wikipedia.org/wiki/Sign_language#Classification
Those relationships might be something that reinforces the impression that all sign languages are inherently related, but there are also sign language isolates that don't have an etymological relationship to other sign languages (although if they've had later contact with other sign languages, they could have loan vocabulary or other contact influences).
https://en.wikipedia.org/wiki/Category:Sign_language_isolate...
Edit: I don't mean to suggest that people who know different sign languages couldn't communicate at all, but I expect that someone who knows one wouldn't be able to understand a signed conversation between native signers of an unrelated one.
If you know sign language, you can pretty much understand what they're saying. The first time I noticed I understood someone in Brazil, talking about work, I didn't notice that... I was understanding them. I didn't understand certain stuff -- slang --.
Yes, abstract and non-literal. It's because of the PC brigade. Growing up... things were a whole lot different. It's still pretty much literal.
https://en.wikipedia.org/wiki/Swadesh_list
Edit: oh, there's already a special Swadesh list for comparing sign languages because of "overestimation of the relationships between sign languages, due to indexical signs such as pronouns and parts of the body". (Those signs are the ones that are most likely to be shared because they're likely to arise independently or have an inherently obvious meaning.)
Navarro wrote a book called "What Every Body is Saying".[0] In it he identifies a number of universal body language patterns and sets out guides for interpreting what they mean based on context.
[0] http://www.amazon.com/What-Every-BODY-Saying-Speed-Reading/d...
i tell them i'm traveling on business, or act vaguely rude to the border agent with very curt responses. apparently they're good at filtering the real liars because i've never been hassled since.
It would be really cool if the liars were rehearsed in the technique.
What is ironic about that? If I understand how to lie to other humans, why is it anything but expected that I would be able to recognize that skill in other people?
Well then take the opposite. What about honest people, do they recognize honest people better? I don't know, just asking, wondering. It could even be that honest people recognize honesty better than liars, while at the same time liars are better lie detectors.
Then there has to be a middle group I guess - what are they?
I also don't really know, but it sounds plausible :)